The overall goal of this proposal is to maintain and expand the population-based, computerized San Francisco Mammography Registry (SFMR) so that it can continue to serve as a resource for conducting high quality, clinically significant research related to breast cancer. Specifically, we will continue to collect demographic, clinical and risk factor information, mammographic interpretations and cancer outcomes obtained through linkage with the regional population-based Surveillance, Epidemiology, and End Results (SEER) program and the California Cancer Registry for the purpose of evaluating the performance of mammography. The cohort of women (N= 178,887) in the SFMR includes 49 percent non- Hispanic whites, 24 percent Asian/Pacific Islanders, 12 percent Hispanics, and 10 percent African Americans. Information on 361,884 mammography examinations has been collected. The number of women will increase over the next five years to an estimated 210,000, and the number of mammograms to an estimated 800,000. We will determine the rate of cancer, sensitivity, specificity and positive predictive value of mammography, and type of cancer detected [DCIS vs. invasive, size, grade, and stage] according to race/ethnicity, risk factors, and screening practices using the SFMR (Research Plan number 1). Additional research objectives (Research Plans number 2-number 6) include: 1) determining whether mammographic breast density and bone mineral density, both strong predictors of breast cancer risk, are correlated, 2) assessing the diagnostic accuracy among radiologists by volume of examinations interpreted per year as well as by physician characteristics using data from the SFMR and three other BCSC registries, 3) utilizing BCSC mammography registries that link with SEER programs to validate a mammography algorithm developed to distinguish screening from diagnostic mammography in the SEER-Medicare database, 4) determining patient and provider characteristics that are associated with inadequate follow-up of women with breast lumps but negative mammography, and 5) examining the prevalence and prognostic value of beta- estrogen and alpha-estrogen receptors among mammographically detected DCIS and early invasive cancers. The SFMR database and research platform will be a valuable resource for addressing issues related to mammography performance, for identifying factors that optimize the quality of mammography, for biologic studies of screen-detected compared with other cancers, for developing clinical guidelines, and for future studies of emergent screening technologies and clinical interventions to improve screening outcomes.